Amit Levy: 2015 Security Workshop

Microcontroller operating systems and frameworks ypically assume that a single, monolithic application will run on an embedded system. Traditionally, optimizing for power, squeezing applications into minimal code and memory allocations, and lacking hardware support have constrained microcontroller applications to be single-function. Newer hardware, however, has changed this paradigm. The microcontroller is growing up and can now support a secure, trusted kernel and multiple, isolated, concurrent, and dynamically-loaded applications, all while operating on the power budgets that originally made this device class feasible. While the hardware support is present, the software ecosystem to capitalize on these advances is lagging behind. To remedy this, we propose Tock, a new embedded operating system design that builds on established operating system principles while adapting to the embedded system environment. Tock exploits memory protection units, advancements in modern systems programming languages, and the event driven nature of embedded applications to allow a core kernel, device specific drivers, and untrusted applications to coexist on a single microcontroller. This new operating system will allow embedded devices to mature beyond program once, deploy once systems and into re-usable, ubiquitous, and reliable computing platforms.

Amit Levy is a PhD student at Stanford University, working with David Mazières in the Secure Computer Systems group and Phil Levis in the Stanford Information Networks Group. His work focuses on building pragmatic, secure systems for the Web, for embedded systems and personal computers that increase flexibility for application developers while preserving end-user control of their private data. His work has been presented at top conferences in the fields of operating systems, security and programming languages, has been covered by the New York Times and other media and made it's way into large scale production systems. He holds a BSc in Computer Science and Economics and an MSc in Computer Science from the University of Washington.